- Eric Boahen*
- Department of Statistics School of Sciences, Mathematics and Technology Education University of Technology and Applied Sciences, Ghana
- DOI: 10.5281/zenodo.18710178
Diabetes is a chronic disease that arises when the
pancreas fails to produce sufficient insulin or when the body is unable to
effectively utilize the insulin it produces.
It is one of the major causes of premature illness and
non -communicable diseases accounting for 60% of all death worldwide. This
research was undertaken to unfold the current pattern of diabetes cases in the
Kumasi Metropolis and to come out with an adequate model for forecasting future
trends. Therefore, the research questions include; what is the trend and an
adequate model for forecasting future diabetes cases in Kumasi Metropolis? A
seven year (2006-2013) secondary data was used for this study. .It was sourced
from the Amen scientific herbal clinic with which a time series trend analysis
was used to perform the trend analysis using Minitab version 16. The model
developed from quadratic trend was Y(t)=36.78+0.0275*t-0.00354*t2. Analysis of
the data was carried out using the Box – Jenkins ARIMA (ARIMA) approach with
the help of Gretl.
The presence of unit root was tested using the ADF
test on the time series data from the year 2006 – 2013. The p – values of the
ADF showed the stationarity of the data therefore data is good for ARIMA. This
research was undertaken with the prior motive to develop an adequate model for
forecasting future trends. It evolved that ARIMA (1, 1,1) fits the data well.
We therefore recommend that it should be used in
forecasting future cases of this disease.
